Question 7 of 7: Manufacturing Route Selection for a Door Handle

Question 7: Manufacturing Route Selection for a Door Handle (15 marks)

Question text not reproduced: the examination questions are © Engineers and Geoscientists BC. Open the official past paper (linked at the top of this page) to read the question, then follow the worked solution below.

[Figure not reproduced: Figure 7.1 — the lever handle of the question, redrawn: a rose carrying the spindle boss, and a lever that tapers from the boss to a curved tip. The function is to convert a hand force into a torque on the latch spindle. See the official exam paper.]

Part A — three ways of manufacturing a part with this function

Route 1: hot-chamber pressure die casting in zinc alloy, followed by vibratory finishing and electroplating. Zinc alloys (ZAMAK 3 or 5) cast at low temperature, so tool life is long, and the process produces the tapered, curved, variable-section form of the figure directly, to close tolerance and with a surface fine enough to plate with only vibratory deburring in between. The spindle bore is cast in and reamed. Cycle times are a few seconds, unit cost is low, but the tool is expensive and the design is frozen once it is cut. This is the route behind almost every mass-market residential lever handle.

Route 2: hot forging (stamping) of brass, followed by CNC machining and polishing. A brass billet is heated and closed-die forged to a near-net lever, trimmed, then machined for the spindle bore and the set screw, and polished and lacquered or plated. The result is solid, dense and heavy in the hand, with excellent corrosion resistance and a grain flow that follows the lever, so mechanical performance at the boss is superior to a casting. Tooling is considerably cheaper than a die-casting tool but the material and the secondary machining make the unit cost higher. This is the route for premium architectural hardware. A close relative for stainless steel is investment casting, which buys similar geometric freedom at a higher unit cost and much lower tooling cost.

Route 3: fabrication from formed stainless tube or pressed sheet, welded to a machined boss. A 304 stainless tube is mandrel-bent to the lever profile (or two sheet pressings are drawn and seam-welded to make a hollow lever), then welded to a turned boss and the assembly is ground and brushed. Tooling is a bending mandrel and simple fixtures, so the fixed cost is very low and a design change is cheap; the part is hollow and therefore light and material-efficient, and the corrosion resistance is excellent. Unit cost is high because the labour content is high and the welds must be dressed. This is the commercial and institutional route, and the right answer at low volume.

A fourth route worth naming is injection moulding of a glass-filled polyamide lever over a steel insert at the boss, which is the cheapest of all in high volume and is used for interior residential and appliance handles, but it is normally rejected where perceived quality or fire performance matters.

Part B — factors important in choosing the final process

Production volume against tooling amortisation. This is the dominant factor and it is the reason the same part is made three different ways in three different markets. A process with a high fixed cost and a low variable cost wins above a break-even volume and loses badly below it.

Material, and whether the process can work it. The material is usually chosen for corrosion resistance, strength, density and perceived quality, and it immediately eliminates whole process families — zinc die-casting alloys are not forged commercially, stainless cannot be hot-chamber die cast, brass cannot be injection moulded.

Geometry: section thickness, uniformity, undercuts, hollowness and size. The tapered, curved lever suits casting and forging naturally; making it hollow rules out solid casting; a severe undercut would demand a side action or a different process altogether.

Tolerance and surface finish achievable, and the secondary operations implied. The spindle bore needs a fit tolerance that no casting process holds directly, so a machining operation is implied in every route and must be costed. The visible surfaces must be fine enough for the intended finish, since polishing or plating a rough casting is more expensive than casting it smooth.

Mechanical performance and life. The lever is a cantilever loaded in bending at the boss and cycled tens of thousands of times, and it must survive an abuse load far above the operating load. The process determines the microstructure, the porosity and the residual stress that decide whether it does — forged brass and cast zinc behave quite differently at the boss fillet.

Cost structure, lead time and capacity. Not only the unit cost but its split between fixed and variable, the tooling lead time (often the longest single item in the launch schedule), the supplier's available capacity, and the cost of a design change after the tool is cut.

Supply base, location and risk. Whether a capable supplier exists within an acceptable distance, whether the tooling is single-source, what the transport cost and duty on a heavy part are, and what happens if that one supplier fails.

Regulatory, standards and environmental requirements. Accessibility standards limit the operating force and the shape (CSA B651 requires hardware operable without tight grasping or twisting); fire-rated assemblies restrict materials; plating processes carry effluent and hexavalent chromium restrictions that may exclude a route on environmental grounds alone.

Part C — framework of a selection method

The framework is a four-stage funnel, and its logic is that the cheap, absolute tests are applied first to a wide field and the expensive, relative ones last to a narrow one.

Stage 0 — state the requirement. Extract from the design the attributes a process must deliver: material class, mass and section thickness, overall size, tolerance and surface roughness on each surface, mechanical duty, annual volume and its uncertainty, target unit cost, and the market the product is aimed at.

Stage 1 — screen on hard capability. Pass every candidate process through the attribute limits, exactly as Ashby's process-selection charts do: a process that cannot work the material, cannot reach the section thickness, cannot hold the tolerance or cannot achieve the surface finish is eliminated outright and is not scored. Screening is absolute, uses published process capability data, and costs nothing but an afternoon.

Stage 2 — rank the survivors on cost. Model the unit cost of each surviving process over the volume range, using the standard structure

in which $C_m$ is the material cost per part, $T$ the dedicated tooling cost amortised over the production quantity $N$, $L$ the machine and labour rate and $\dot n$ the production rate, with the last two terms customarily collected into a single variable cost per part. The intersections of these curves are the break-even volumes that decide the answer.

Given. For the lever handle at the design stage, three surviving routes with the following estimates: die casting, tooling CAD 48 000 and variable cost CAD 3.10 per part; brass forging, tooling CAD 22 000 and CAD 5.40 per part; welded stainless fabrication, tooling CAD 3 000 and CAD 12.00 per part.

Find. The volume ranges over which each route is the cheapest, so that the choice can be made against the marketing forecast.

1. Write the break-even condition between two routes. Two processes cost the same when their total costs are equal, so with tooling $T$ and variable cost $u$, $$T_1 + u_1 N = T_2 + u_2 N \quad\Longrightarrow\quad N^{*}=\frac{T_2-T_1}{u_1-u_2}$$ where route 2 is the one with the higher tooling and the lower variable cost.
2. Fabrication against forging. Substituting $T_1 = 3\,000$, $u_1 = 12.00$, $T_2 = 22\,000$, $u_2 = 5.40$, $$N^{*}=\frac{22\,000-3\,000}{12.00-5.40}=\frac{19\,000}{6.60}=2\,879\ \text{parts}$$ so below roughly 2 900 parts a year the welded fabrication is cheapest.
3. Forging against die casting. Substituting $T_1 = 22\,000$, $u_1 = 5.40$, $T_2 = 48\,000$, $u_2 = 3.10$, $$N^{*}=\frac{48\,000-22\,000}{5.40-3.10}=\frac{26\,000}{2.30}=11\,304\ \text{parts}$$ so forging is cheapest between about 2 900 and 11 300 parts a year.
4. Check the envelope with a direct evaluation. At $N=5\,000$ the three unit costs are $3\,000/5\,000+12.00=12.60$, $22\,000/5\,000+5.40=9.80$ and $48\,000/5\,000+3.10=12.70$ CAD, confirming that forging governs in the middle band; at $N=20\,000$ they are 12.15, 6.50 and 5.50 CAD, so die casting governs above the upper break-even. $$\boxed{N < 2\,900:\ \text{welded fabrication};\quad 2\,900 \le N < 11\,300:\ \text{brass forging};\quad N \ge 11\,300:\ \text{zinc die casting}}$$

Stage 3 — score the survivors on the criteria cost does not capture. Apply the weighted-scoring method of question 1 to the remaining routes on quality and consistency, tooling lead time, flexibility to a later design change, supplier risk and proximity, environmental burden (plating effluent, recycled content, transport) and perceived quality in the target market. Cost enters this matrix as one criterion among several, never twice.

Stage 4 — buy the information the decision still lacks, then record it. Obtain firm quotations, commission sample parts from the two leading routes, test them against the mechanical and finish requirements, and qualify the supplier. Then write the decision, its criteria, its weights, its sensitivity and the volume assumption into the design record, because the volume assumption is the thing most likely to change and the record is what allows the decision to be revisited rather than re-argued.